Soils and Foundations 2014;54(6):1039–1053

HOSTED BY The Japanese Geotechnical Society

Soils and Foundations

www.sciencedirect.com
journal homepage: www.elsevier.com/locate/sandf

Quality control of double fluid jet grouting below groundwater

table: Case history
James C. Nia, Wen-Chieh Chengb,n,1
a
Department of Civil Engineering, National Taipei University of Technology, Taipei 10608, Taiwan, ROC
b
College of Engineering, National Taipei University of Technology, Taipei 10608, Taiwan, ROC
Received 16 October 2012; received in revised form 30 July 2014; accepted 14 August 2014
Available online 22 December 2014

Abstract

A jet grout block created by overlapping columns aimed to convert the saturated silty sand soils into a safe space for breaking down four
underground parking uplift piles along the tunnel alignment. Since the grouting platform was located on the mat foundation in the basement of
underground parking and was under the groundwater table, a series of dewatering wells were installed before the 177 columns were constructed
to prevent intrusion of the groundwater flowing upward from sandy soil to basement through grouting hole during wash boring and grouting
stages from happening. Several quality control measures were undertaken prior to quality assurance testing. Two measurements of spoil return
and spoil flow rate for each column were implemented. From a back analysis from the mean values of spoil density and spoil flow rate, the
column diameter was estimated at around 1.56 m, slightly smaller than the design diameter by 2.5%. In addition, a control chart with upper and
lower control limits, established from a large dataset of flow rate of spoil return, was a means to recognize the likelihood of sand boiling and to
serve as an early warning indicator of abnormal conditions for the jet grouting works below the groundwater table. As the mean spoil density of
the infill column was larger than perimeter column, an optimal grouting sequence for clusters was suggested in this paper. The jet grout block
exposed during the breakdown process of the piles showed safe working conditions as required and a shield machine then passed through as
planned.
& 2014 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved.

Keywords: Double fluid jet grouting; Spoil density; Spoil flow rate; Column diameter; Infill column

1. Introduction
There are three jet grouting systems to choose from: a
single, double, and triple system. The single system injects
grout only at high pressure. This was the first system to be
n
Corresponding author. Tel.: þ886 2 2771 2171x2627;
fax: þ886 2 2781 4518.
E-mail addresses: ckni@ntut.edu.tw,
s2428030@gmail.com (W.-C. Cheng).
1 friction loss, allowing the cement grout to travel further from
Contact address: 1, Section 3, Chung-hsiao E. Rd., Taipei 10608, Taiwan,
ROC.
Peer review under responsibility of The Japanese Geotechnical Society.
used: the resulting column diameter is limited and the bore-
holes sometimes become jammed, resulting in ground heave
(Croce and Flora, 1998). The column sizes are small, usually
less than 1 m in diameter.
The JSG (Jumbo-jet Special Grout) method is a jet grouting
method with two fluids, neat cement grout and air (Yahiro
et al., 1982; Miki and Nakanishi, 1984). The cement grout is
injected at high pressure and is aided by a cone of compressed
air, which shrouds the grout injection. The air reduces the
the injection point, thereby producing larger column diameters.
The addition of the air shroud increases jetting efficiency

http://dx.doi.org/10.1016/j.sandf.2014.11.001
0038-0806/& 2014 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved.
1040 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053
dramatically, typically resulting in an increase of 30% or in the
design diameter for the same jetting energy. Nevertheless, the
presence of the air bubbles means that that the column is not as
strong as those produced by the single-fluid method, and more
spoil return is produced.
The triple system differs from the single and double systems
in that the erosion of the ground is carried out by a high
pressure water jet shrouded with air and the eroded region is
replaced by cement grout with an additional low pressure grout
line. The single and double fluids are in-situ mixing methods,
not soil replacement methods like the triple fluid method.
As mentioned above, the double fluid jet grouting technique
results in the partial replacement of in-situ soil within the
grouted column, thereby improving strength and reducing
permeability. Because the overlap of the column with pre-
viously constructed columns can be accurately controlled, low
permeability can be secured, thus avoiding the risk of leakage
of water or soil through the gap between jetted columns. The
accuracy of any quality control technique to evaluate the actual
diameter of jet grout column is thus of major concern to the
grouting industry.
The jet grouting diameter is affected by soil properties and
jet grouting parameters (the lifting speed and rotation fre-
quency of the monitor, the grout volume, the grout pressure,
and the nozzle diameter) (Malinin et al., 2010). Therefore,
these parameters are required for manually monitoring and for
the automatic data acquisition system. However, either of these
procedures provide solid values with regard to improved
strength and the diameter of jet column. Quality assurance
(Wang et al., 2013) is therefore generally adopted by sampling
grout cubes and spoil cubes and performing breaks on both,
along with taking spoil density readings throughout the jet
grouting process to check the strength compatibility with the
trial test sections (Langhorst et al., 2007; Hurley, 2005; Stark
et al., 2009). Similarly, Kauschinger et al. (1992) developed
mass balance equations for the single fluid jet grout system,
which can be adopted to estimate the size and composition
(soil, water, and cement) of spoil and soilcrete (jet grout
column) by measuring the densities of soilcrete, cement grout,
and spoil.
Hydrophones were used by Langhorst et al. (2007) and
Hurley (2005) to detect the vibrations in the tubes at defined
radii during the trial grouting program. This gave qualitative
indications that the desired column diameter was being
achieved which, when combined with a back analysis of spoil
return densities, provided the confidence to proceed with
production work. However, the installation of tubes for
hydrophones is costly and time consuming. This will limit
hydrophone method only to trial grouting programs, and
makes it impractical for quality control during production
work using the jet grout column.
Langhorst et al. (2007) and Malinin et al. (2010) used an
innovative measuring device which was made of two crossing
bars moving apart against the wall of the column wall before
the grout set. Although the measured diameters of the columns
were well-matched with those observed from the exposed
columns after excavation, this device is not effective with jet
columns at greater depths. In addition, Ho (2011) used an
analytical model to predict the cutting distance of air-shrouded
jets in cohesive soil. The model variables were calibrated by
back-analyzing jet grouting field trial results at a nearby site
with similar soil conditions.
Flora et al. (2012) proposed a most promising jet grouting
design method by conducting a statistical analysis to obtain the
mean value and coefficient of variation of the column diameter
and verticality based on the measured values from a jet-
grouting field trial. Then the uplift safety factor of the water
sealing barrier made by jet grouting at the bottom of open
excavation (Croce and Flora, 2000; Croce and Modoni, 2007;
Flora et al., 2007; Lignola et al., 2008; Modoni et al., 2006)
was estimated by the Monte Carlo procedure by assuming the
column diameter, column spacing, and a given risk level. This
method is quite simple and also accurate, and is ideal for use as
a quality control method for production work.
This study presents a case history where a double fluid jet
grouting project was conducted to maintain safety when four
underground parking uplift piles were broken (Xu et al., 2014)
during subway tunnel construction. Since the grouting plat-
form was below the groundwater table, there was a possibility
that groundwater would flow upward from the sandy soil to the
basement through the grouting hole during the wash boring
and grouting stages. Such leakage may have led to a serious
incident, such as sand boiling. Therefore, two measurements
were taken: spoil return and the spoil flow rate for the soilcrete
column. A back analysis from the mean values of spoil density
and spoil flow rate was utilized to estimate the column
diameter. A control chart with upper and lower limits was
developed in this paper to provide an early warning indicator
of water leakage from a sand boiling issue. Apart from that, an
optimal grouting sequence in a cluster was suggested in this
paper in order to have more infill columns than perimeter
columns.
2. Site characterization
Two 6.5-m diameter bored parallel tunnels driven using
Earth Pressure Balance (EPB) shields need to pass beneath an
existing underground parking under where four uplift piles,
indicated by the red circle in Fig. 1, are in the way of the
alignment. Double fluid jet grouting at depths ranging from
23.27 to 34.91 m was undertaken to improve the strength and
reduce the permeability of the silty sand layer below the mat
foundation of the underground parking. The purpose of this
exercise is to provide a safe place for excavating the soils
around the piles and then removing these piles from the
alignment (Figs. 1 and 2). To prevent substantial structure
heaves or settlements from occurring during jet grouting, real-
time monitoring was implemented using electronic beam
sensors installed in the underground parking (Fig. 1).
As seen in Fig. 2, the soil succession at this construction site
of Taipei City comprises a fill of about 1.35-m in thickness,
sitting on a 2.9-m thick medium stiff silty clay underlain by
11.15-m thick loose silty sand then 7.35-m thick medium stiff
silty clay which extends to a thick layer of medium dense silty
 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053 1041

N
Southbound
tunnel Northbound
tunnel
Diaphragm
wall
4.25
LEGEND 4.25

JSG grouting
Double tube grouting
TAM grouting BP4
Pumping well
BP3
Surface settled area
Point well
Surface grout hole Under-
Stair 8.00 BP2 ground
Sand boiling parking
Electronic beam sensor lot
Pile to be removed
8.00 BP1
Soilcrete
column
(surface)

Unit:
meter

Fig. 1. Plan view of the four piles in way of tunneling.

sand, followed by a gravel formation. The water table of this
11.15-m thick silty sand layer is some 4.25 m below the
surface and the water table of the thick lower silty sand layer,
with a thickness of 19.25 m, is 6.7 m deep.
Since the top sand layer under the mat foundation was
laterally enclosed by a diaphragm wall and underlaid by an
impermeable clay layer, it was presumed that the groundwater
of this sand layer within this confined region was hydraulically
cut off from the same soil layer outside (Fig. 2). Therefore, the
groundwater in this top sand layer would not flow into the
basement through grouting holes unless there were water
seepage paths in diaphragm wall. Nevertheless, because it
was possible that the groundwater level in the bottom sand
layer, which was higher than the grouting platform, could flow
into the basement through grouting holes, a dewatering system
was installed before the grouting holes were drilled. To avoid
possible surface settlement and building tilting outside the
diaphragm wall due to the lowering of the groundwater level,
dewatering wells were installed to the right of the grouting
area, far away from the diaphragm wall. A total of six point
wells and three pumping wells were deployed evenly into three
pits, as shown in Fig. 1. The point wells were opened to lower
the groundwater level in the top sand layer to just below the
mat foundation and the pumping wells were then opened to
maintain the groundwater level in the bottom sand layer at a
similar level.
3. Jet grouting work
A total of 187 jet columns1.6 m in diameter were con-
structed to provide a safe working space for breaking down the
uplift piles inside the jet grout block with a minimum
unconfined compressive strength of 2 MN/m2 for sandy soils
or 1 MN/m2 for clayey soils and a permeability of less than
1  10  7 m/s. A 1000 l grout mix consisting of 120 kg of type
1 Portland cement with a specific gravity of 3.15, 280 kg of fly
ash with a specific gravity of 2.86, and 864 kg of water
provided a density of 1.264  103 kg/m3 of cement-based
grout to be used for double fluid jet grouting.
1042 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053

Surface Diaphragm
settled area wall
EL. 103.70 m
BF
N=6 EL. 102.35 m
CL γ=19.0 kN/m3 Basement
EL. 99.45 m
ϕ=31°
EL. 99.45 m

EL. 97.00 m Point Pumping

0.20 0.20 well well
N=8 EL. 94.70 m
SM γ=19.4 kN/m3 Eroded
area EL. 93.60 m
ϕ=32°
Inflow
Void
of wall BP1 Screen
EL. 88.30 m Casing

N=7
3
CL γ=18.9 kN/m
ϕ=31° 4.04
1.50
Submersible
EL. 80.95 m
N=18 pump
3
SM/ML γ=19.5 kN/m
ϕ=32° Southbound
EL. 76.70 mtunnel
N=18 Northbound
SM/ML γ=19.5 kN/m3 tunnel
ϕ=32°
EL. 72.70 m Screen

Grout block

N=27
SM γ=19.8 kN/m3 2.10 6.24 2.43
ϕ=33° 10.77

EL. 61.70 m
Fig. 2. Subsurface profile along with grout block and construction details of the wells.

As shown in Fig. 1, 177 columns were installed from the Table 1

mat foundation floor within this underground parking structure Jet grouting parameters for production work.
and 10 columns, marked by the green circle, were installed at
Air pressure (MN/m2) 0.7
the ground surface outside the structure. Due to the constraint Air flow rate (l/min) 1500
of the working area imposed by the presence of existing beams Grout pressure (MN/m2) 20 72
and columns on the mat foundation, jet grout holes were Grout flow rate (l/min) 60
drilled from within pits surrounded by beams. While most Withdrawal rate (min/m) 30
columns were drilled vertically, for some columns, the drilling Rotation speed (rpm) 6 to 8
Water–cement ratio 2.16
inclination was up to 20 degrees from the vertical. At each
grouting pit, a cluster of 13 to 15 columns were constructed to
form overlapping grout columns, as shown in Fig. 1. grout, but also to provide soil lift. The jetting parameters
At each column position, a borehole was first drilled to the summarized in Table 1 are justified by the design chart of
design depth and then the jetting monitor was withdrawn at a Japanese Jet Grouting Association (JJGA, 2002) based on the
speed of 30 min/m and rotated with a speed of 6 to 8 rpm, soil type, the SPT blow count, and depth, and were adopted for
while cement grout was grouted under pressure of 20 MN/m2 the production grouting work in this project.
through the nozzle of 2.8 mm in diameter at 0.06 m3/min and
an air flow of 0.06 m3/min was injected around the jet grout 4. Volume and density measurements of spoil return
with a pressure of 0.7 MN/m2. The pressure and flow rate of
the jetting fluids was sufficient to erode the soil and to mix it. The jet-grouting spoil was discharged through annulus
Pressurized air was used not only to conserve the energy of jet between drilling rod and pipe. In this project, the spoil was
 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053
Table 2
Ratio of the spoil volume to the grout volume for a unit length of column.
Column no. Measured section (from tip) Ratio
Machine no.1
4 1 1.49
8 1.37
5 1 1.82
2 1.62
3 1.43
9 1.37
18 2 1.37
5 1.40
24 2 1.22
8 1.06
23 1 1.77
2 1.38
5 1.24
22 1 1.67
2 1.34
8 1.24
21 1 1.92
2 1.52
9 0.86
10 1.38
8 1 1.56
2 1.33
8 1.37
9 2 1.37
8 1.37
10 1 0.93
9 1.01
14 1 1.06
8 1.06
15 2 1.62
9 1.47
19 1 1.01
9 1.17
20 1 1.17
8 1.22
25 2 1.11
8 1.39
28 2 1.11
7 1.01
29 2 1.47
9 1.42
30 1 1.62
2 1.40
7 1.29
31 1 1.17
8 1.22
32 2 1.26
7 1.22
33 2 1.82
3 1.37
7 1.17
34 2 1.62
3 1.30
8 1.22
35 2 1.87
3 1.45
6 1.26
36 3 1.47
8 1.37
37 2 1.06
6 1.37
1043
Table 2 (continued )
Column no. Measured section (from tip) Ratio
38 2 1.56
3 1.39
7 1.47
39 2 1.22
8 1.77
9 1.41
44 1 0.97
8 1.37
46 1 1.17
7 0.91
55 4 0.97
9 1.31
65 2 1.26
6 1.23
66 2 1.01
9 1.46
Avg. 1.33
Machine no.2
43 1 1.22
9 1.47
51 2 1.01
8 1.22
59 4 1.31
8 1.37
60 2 0.86
3 1.24
8 1.11
67 3 1.11
9 1.11
68 3 1.22
7 1.47
70 2 0.91
3 1.39
8 1.22
71 1 1.26
8 1.26
73 2 1.06
7 1.37
76 3 1.11
8 1.42
77 1 1.17
8 1.11
78 2 1.22
7 1.11
79 2 1.01
8 1.06
80 1 1.11
9 1.22
97 2 0.99
9 1.17
89 1 1.37
10 1.77
88 1 1.44
8 1.22
84 2 1.67
4 1.41
9 1.29
85 3 1.52
9 1.11
86 2 1.22
9 1.17
87 3 1.77
4 1.40
1044 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053
Table 2 (continued )
Column no. Measured section (from tip) Ratio
9 1.37
93 1 1.22
8 1.24
94 2 1.13
10 1.01
95 2 1.06
8 1.37
96 1 1.92
2 1.43
8 1.42
102 3 1.37
7 1.42
103 2 1.06
8 0.86
9 1.19
104 2 1.22
8 1.26
105 2 1.26
8 1.72
9 1.42
106 2 1.06
8 1.11
113 2 1.13
7 1.09
122 1 1.19
2 1.11
Avg. 1.25
Machine no. 3
130 3 1.06
10 1.52
137 4 1.62
5 1.34
8 1.37
138 2 1.47
8 1.31
139 4 1.62
5 1.34
10 1.37
145 1 1.52
3 1.43
10 1.06
146 3 1.37
7 1.31
147 3 1.37
9 1.31
148 3 1.52
4 1.40
10 1.22
155 1 1.87
2 1.44
8 1.37
156 2 1.26
9 1.37
157 4 1.11
8 1.13
163 1 0.86
3 1.18
9 1.17
164 3 1.22
8 1.37
165 2 1.72
3 1.39
Table 2 (continued )
Column no. Measured section (from tip) Ratio
8 1.22
166 4 1.11
9 1.22
174 3 1.37
7 1.31
175 3 1.47
8 1.31
181 2 1.31
8 1.29
182 3 1.26
9 1.42
183 2 1.28
7 1.23
187 2 1.17
9 1.17
Avg. 1.33
Summary
Machine no. Avg. ratio
Machine 1 1.33
Machine 2 1.25
Machine 3 1.33
Total avg. 1.30
collected in a confined temporary holding space and then
conveyed to a pit of 3.79 m (L) by 2.40 m (W) by 1.25 m (H)
where the spoil volume was measured before it was pumped to
the ground surface to be shipped out. The density of spoil
returns sampled from the grout hole was also measured using
the mud balance along with the measurement of spoil flow
volume at 1 m intervals of during jet grouting. To ensure their
validity, measurements were taken twice for each column, one
2 to 4 m from the column tip and the other 7 to 11 m from the
column tip.
5. Volume measurement of spoil return
As cement grout is jetted into saturated in-situ soil to create
a mixture of grout and soil, some remains in the ground to
form a jet grout column and some flows to the surface. The soil
which returns to the surface is referred to as spoil return.
According to the theory of mass balance, the volume of jetted
grout should be equal to the volume of spoil return (mixture
flowing to the ground surface). However, in reality the volume
of spoil return is larger than the volume of jetted grout, and the
amount of depends on the soil type, such as clay or sand. The
design value of the volume ratio of spoil to grout is suggested
by JJGA is 1.3 when using clayey soil, and 1.1 for sandy soil.
Since the grouted area in this construction site consists of silty
sand mixed with layers of silty clay, the average volume ratio,
calculated by the spoil return volume divided by the grout
volume for 1 m long column, was expected to be in the
1.1 to 1.3 range. The grout volume injected during 30 min of
lift for a 1 m long column with an injection rate of 60 l/min is
1.8 m3, and the spoil volume is measured in a pit with known
 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053 1045

length and width dimensions. The ratios of spoil returns
to grout volumes measured for a unit length of columns are
summarized in Table 2 and can be also presented as a control
chart in the order of production sequence, as shown in Fig. 3.
With a mean value of 1.302, the upper control limit (mean
value plus 3 times the standard deviation) and lower control
limit (mean value minus 3 times the standard deviation) are
determined to have been 1.970 and 0.634, respectively. This
slightly higher mean value of 1.302, not between 1.1 and 1.3
for silty sand mixed with layers of silty clay suggested by
JJGA, could be caused primarily by the addition of ground-
water into spoil as grouting platform is below groundwater
table. The other possible cause may be the expansion of soil
particles due to the effective stress relief of in-situ soil under

LEGEND
Before sand boiling
Machine No.1
Machine No.2
Machine No.3

0 20 40 60 80 100
2.8 2.8

2.4 2.4
of spoil to grout
Volume ratio

1.970 (Upper control limit)

2 2

1.6 1.6
1.302 (Mean value)
1.2 1.2

0.8 0.634 (Lower control limit) 0.8

0.4 0.4
1.6

0.8
0.4
2

1.2

0 20 40 60 80 100
Probability Density Function Sequence of production work
PDF
Fig. 3. Control chart of the volume ratio between the spoil return and grout.

Fig. 4. (a) Photo of a pit near diaphragm wall before sand boiling, (b) a hole left by withdrawing intermediate pile in the pit, and (c) sand boiling in a pit next to the
grouting pit.
1046 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053

Fig. 5. (a and b) Pits flooded with boiling sand and (c) a surface settled area.

underground parking. Any possible random errors from
measurement should be dealt with statistical analysis.
According to the Western Electric Company (1956), a
process is considered out of control if the following criterion
is met: Two of the three most recent points plot outside and on
the same side as one of the 2-sigma control limits. Juran (2010)
also proposes the same criteria. The chance that two out of
three points in a row fall outside the warning limit is only
about 1%. In Fig. 3, the volume ratios of spoil to grout for the
first five columns were 1.24, 1.31, 1.81, 1.95, and 2.65 shown
by cross points. The three most recent ratios are bigger than
the 2-sigma control limit (mean value plus 3 times standard
deviation), which is 1.746. At this time, sand boiling (Fig. 4c)
from a hole in the pit was observed where jet grouting near
diaphragm wall (Figs. 1, 4a and b) was conducting. It is
believed that this square hole was the space left out unfilled
after the basement structure was completed and intermediate
piles were withdrawn along with bracing system in the past.
The flow path of spoil return eroded the soils away between
the hole in the pit and the opening on diaphragm wall (Fig. 2)
which was caused by inadequate concreting (Ni and Cheng,
2012). The groundwater started to flow through the opening
from outside the diaphragm wall because the water level was
higher than the grouting pit. This resulted in an increase in the
 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053 1047

volume ratios of spoil to grout for the first five columns and Table 3
ultimately triggered the sand boiling (Fig. 5a and b) for several Density of the spoil return.
minutes and an extensive surface settled area outside the Column no. Density of spoil
diaphragm wall, as shown in Figs. 1, 2 and 5c, until the water (103 kg/m3)
levels on both sides of wall were equalized. Top
It is obvious that a control chart with upper and lower control Bottom
limits established from a large r dataset of flow rate of spoil
4 1.45
return along with proper decision-making criteria would provide 1.48
a way to identify the potential of sand boiling and would serve 5 1.39
as an early warning indicator of abnormal conditions for the jet 1.40
grouting works below the groundwater table. In addition, sand 8 1.40
1.37
boiling events may be avoided by installing casing from the 9 1.47
surface in the pit to the top elevation of the jet grout column 1.43
(Fig. 2) before jetting as was recommended by Ho (2011). This 10 1.37
not only would keep the spoil flow path free from potential 1.40
blockage, but also eliminate the possibilities of soil erosion and/ 14 1.40
1.37
or sand boiling.
15 1.36
1.40
6. Density measurement of spoil return 18 1.40
1.41
The densities of spoil returns defined as the weights per unit 19 1.49
1.49
volume of spoil returns and measured for a unit length of column
20 1.34
are listed in Table 3 and are also plotted as a control chart in the 1.44
order of production sequence, as shown in Fig. 6. The mean 21 1.40
value is 1.409  103 kg/m3, and the upper control limit and lower 1.30
control limit are 1.585  103 kg/m3 and 1.232  103 kg/m3, 22 1.43
1.36
respectively. The values of standard deviation and coefficient of
23 1.36
variation are 0.059 and 0.001, respectively. 1.32
In an ideal situation, if the jet grout mixes completely with
34 1.36
the in-situ silty sands with several layers of silty clay (Croce
1.44
and Flora, 2000), then the composition in the jet grout column 35 1.45
and in the spoil return should be the same. Due to the effect of 1.28
bleeding and consolidation, the column density should even- 36 1.37
tually be larger than the spoil density. Hurley (2005) proposed 1.49
37 1.30
a quality assurance principle by sampling grout cubes and spoil
1.31
cubes and performing breaks on both, along with taking spoil 38 1.35
density readings throughout each jet grout lift to measure 1.28
compatibility with the trial test sections. 39 1.48
Herein, the question is how the spoil density is related to the 1.37
43 1.34
column diameter. Field borehole loggings indicated the aver-
1.41
age density of in-situ soil is 1.95  103 kg/m3, which is larger 44 1.47
than that of grout; that is, 1.264  103 kg/m3 (Table 4). 1.37
Therefore, the density of the spoil should be in the range of 46 1.56
1.95  103 kg/m3 and 1.264  103 kg/m3. If more in-situ soils 1.34
51 1.62
are eroded and mixed with cement grout to form a jet grout
1.48
column, the density of the spoil return should be on the high 55 1.37
end. Otherwise, the density of spoil will be lower if the spoil 1.43
return volume is bigger than the jet grout volume. With this 59 1.40
relationship between the spoil density and spoil return volume 1.39
60 1.56
in mind, the following section will present a simple method to
1.39
estimate the column diameter by means of a back analysis
from the mean values of the spoil density and spoil flow rate. 77 1.39
1.38
78 1.52
7. Estimation of column diameter 1.49
79 1.48
Spoil densities of column diameters ranging from 1 to 2.6 m 1.53
with 1 m long are listed in Table 4, where the total weights of 80 1.42
1048 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053

Table 3 (continued ) Table 3 (continued )

Column no. Density of spoil Column no. Density of spoil

(103 kg/m3) (103 kg/m3)
Top Top
Bottom Bottom

1.37 1.41
84 1.41 33 1.40
1.43 1.40
85 1.59
1.51 65 1.29
86 1.44 1.31
1.52 66 1.52
87 1.43 1.42
1.34 67 1.36
88 1.52 1.34
1.40 68 1.42
89 1.46 1.42
1.32 69 1.44
93 1.34 1.34
1.39 70 1.32
94 1.46 1.46
1.49 71 1.32
95 1.49 1.36
1.47 73 1.37
1.44
137 1.42 76 1.42
1.39 1.38
138 1.42
1.32 96 1.56
139 1.46 1.50
1.45 97 1.38
145 1.42 1.30
1.41 102 1.44
146 1.38 1.39
1.39 103 1.42
147 1.37 1.45
1.34 104 1.38
148 1.36 1.36
1.40 105 1.41
155 1.33 1.47
1.28 106 1.41
156 1.42 1.43
1.33 113 1.38
157 1.36 1.30
1.42 130 1.36
163 1.41 1.40
1.43 166 1.32
164 1.47 1.36
1.44 174 1.32
165 1.44 1.36
1.42 175 1.43
24 1.51 1.39
1.47 181 1.41
25 1.42 1.37
1.31 182 1.44
26 1.42 1.50
1.29 183 1.37
27 1.37 1.40
1.52 187 1.52
28 1.45 1.43
1.39 Avg. 1.41
29 1.42
1.39
30 1.41 the in-situ soil with the same volume of jet grout column and
1.34 cement grout are divided by the total volumes of the jet grout
31 1.36 column and the cement grout if the jetted grout is thoroughly
 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053 1049

LEGEND
Top of column
Bottom of column

0 20 40 60 80 100
1.7 1.7

Density of spoil return

1.6 1.585 (Upper control limit) 1.6

(103kg/m3)
1.5 1.5

1.409 (Mean value)

1.4 1.4

1.3 1.3
1.232 (Lower control limit)
1.2 1.2
8

0 20 40 60 80 100
Probability Density Function Sequence of production work
PDF
Fig. 6. Control chart of the density of the spoil return.

Table 4
Variation of the spoil density and column diameter.

Column Height Column In-situ soil density Grout volume Grout density Spoil density Adjust Adjusted spoil density
diameter (m) (m) volume (m3) (103 kg/m3) (m3) (103 kg/m3) (103 kg/m3) factor (103 kg/m3)

1 1 0.79 1.95 1.8 1.264 1.47 1.210 1.217

1.1 1 0.95 1.95 1.8 1.264 1.50 1.198 1.253
1.2 1 1.13 1.95 1.8 1.264 1.53 1.186 1.289
1.3 1 1.33 1.95 1.8 1.264 1.56 1.174 1.325
1.4 1 1.54 1.95 1.8 1.264 1.58 1.163 1.359
1.5 1 1.77 1.95 1.8 1.264 1.60 1.152 1.392
1.6 1 2.01 1.95 1.8 1.264 1.63 1.143 1.422
1.7 1 2.27 1.95 1.8 1.264 1.65 1.134 1.452
1.8 1 2.54 1.95 1.8 1.264 1.67 1.125 1.481
1.9 1 2.83 1.95 1.8 1.264 1.68 1.117 1.507
2.0 1 3.14 1.95 1.8 1.264 1.70 1.110 1.532
2.1 1 3.46 1.95 1.8 1.264 1.72 1.103 1.555
2.2 1 3.80 1.95 1.8 1.264 1.73 1.097 1.577
2.3 1 4.15 1.95 1.8 1.264 1.74 1.091 1.597
2.4 1 4.52 1.95 1.8 1.264 1.75 1.086 1.616
2.5 1 4.91 1.95 1.8 1.264 1.77 1.081 1.634
2.6 1 5.31 1.95 1.8 1.264 1.78 1.061 1.674

Fig. 7. Types of the overlapping columns.

1050 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053

Table 5 Table 5 (continued )

Variation of the volume ratio of spoil to grout and the spoil density due to
overlapping types. Column no. Ratio of spoil Density of spoil
return to grout return (103 kg/m3)
Column no. Ratio of spoil Density of spoil
return to grout return (103 kg/m3) 1.17 1.38
103 1.06 1.45
Type 0 0.86 1.42
30 1.62 1.34 1.19
1.40 104 1.22 1.36
1.29 1.41 1.26 1.38
34 1.62 1.44 113 1.13 1.30
1.30 1.09 1.38
1.22 1.36
55 0.97 1.43 122 1.19 1.43
1.31 1.37 1.11 1.41
65 1.26 1.31 148 1.52 1.40
1.23 1.29 1.40
106 1.06 1.43 1.22 1.36
1.11 1.41 155 1.87 1.42
163 0.86 1.43 1.44
1.18 1.37 1.28
1.17 1.41 166 1.11 1.36
183 1.28 1.40 1.22 1.32
1.23 1.37 174 1.37 1.36
Avg. 1.24 1.39 1.31 1.32
181 1.31 1.37
Type 1 1.29 1.41
5 1.82 1.40 182 1.26 1.50
1.62 1.42 1.44
1.43 Avg. 1.31 1.39
1.37 1.39
10 0.93 1.40 Type 2
1.01 1.37 4 1.49 1.48
15 1.62 1.40 1.37 1.45
1.47 1.36 21 1.92 1.30
18 1.37 1.41 1.52
1.40 1.40 0.86 1.40
20 1.17 1.44 1.38
1.22 1.34 23 1.77 1.32
32 1.26 1.38 1.38
1.22 1.43 1.24 1.36
33 1.82 1.40 25 1.11 1.31
1.37 1.39 1.42
1.17 1.40 31 1.17 1.41
36 1.47 1.49 1.22 1.36
1.37 1.37 38 1.56 1.28
37 1.06 1.31 1.39
1.37 1.30 1.47 1.35
39 1.22 1.37 43 1.22 1.41
1.77 1.47 1.34
1.41 1.48 46 1.17 1.34
0.91 1.56
44 0.97 1.37 51 1.01 1.48
1.37 1.47 1.22 1.62
66 1.01 1.42 59 1.31 1.39
1.46 1.52 1.37 1.40
67 1.11 1.34 68 1.22 1.42
1.11 1.36 1.47 1.42
77 1.17 1.38
1.11 1.39 70 0.91 1.46
84 1.67 1.43 1.39
1.41 1.22 1.32
1.29 1.41 79 1.01 1.53
89 1.37 1.32 1.06 1.48
1.77 1.46 88 1.44 1.40
93 1.22 1.39 1.22 1.52
1.24 1.34 94 1.13 1.49
97 0.99 1.30 1.01 1.46
 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053 1051

Table 5 (continued ) Table 5 (continued )

Column no. Ratio of spoil Density of spoil Column no. Ratio of spoil Density of spoil
return to grout return (103 kg/m3) return to grout return (103 kg/m3)

137 1.62 1.39 Type 4

1.34 14 1.06 1.37
1.37 1.42 1.06 1.40
145 1.52 1.41 19 1.01 1.49
1.43 1.17 1.49
1.06 1.42 24 1.22 1.47
156 1.26 1.33 1.06 1.51
1.37 1.42 29 1.47 1.39
175 1.47 1.39 1.42 1.42
1.31 1.43 35 1.87 1.28
187 1.17 1.43 1.45
1.17 1.52 1.26 1.45
Avg. 1.30 1.42 80 1.11 1.37
1.22 1.42
Type 3 87 1.77 1.34
8 1.56 1.37 1.40
1.33 1.37 1.43
1.37 1.40 96 1.92 1.50
22 1.67 1.36 1.43
1.34 1.42 1.56
1.11 1.43 Avg. 1.35 1.43
28 1.11 1.39
1.01 1.45 Type 5
60 0.86 1.39 164 1.22 1.44
1.24 1.37 1.47
1.11 1.56 Avg. 1.29 1.46
71 1.26 1.36
1.26 1.32 Type 6
73 1.06 1.44 9 1.37 1.43
1.37 1.37 1.37 1.47
76 1.11 1.38 Avg. 1.37 1.45
1.42 1.42 Summary
78 1.22 1.49 Overlapping type Volume ratio of Density of spoil
1.11 1.52 spoil to grout return (103 kg/m3)
85 1.52 1.51 Type 0 1.24 1.39
1.11 1.59 Type 1 1.31 1.39
86 1.22 1.52 Type 2 1.30 1.42
1.17 1.44 Type 3 1.30 1.42
95 1.06 1.47 Type 4 1.35 1.43
1.37 1.49 Type 5 1.29 1.46
102 1.37 1.39 Type 6 1.37 1.45
1.42 1.44
105 1.26 1.47
1.72 1.41 mixed with native soil. However, these spoil densities are
1.42 overestimated because the mean ratio of 1.302 of spoil return
130 1.06 1.40
volume to grout volume is larger than 1. An adjusted factor
1.52 1.36
138 1.47 1.32 defined as the ratio of the sum of column volume and spoil
1.31 1.42 volume (1.302 times of grout volume) to the sum of column
139 1.62 1.45 volume and grout volume is suggested in this paper to provide
1.34 a rational estimate of spoil density. The column diameter most
1.37 1.46
likely achieved in the field can be estimated by referring the
146 1.37 1.39
1.31 1.38 measured mean spoil density of 1.41  103 kg/m3 (Table 3) to
147 1.37 1.34 the adjusted spoil density, as summarized in Table 4.
1.31 1.37 This back-calculated diameter of jet grout column is 1.56 m,
157 1.11 1.42 slightly smaller than the design diameter of 1.6 m by 2.5%.
1.13 1.36
Overall, the scatter in the column diameters reflects the natural
165 1.72 1.42
1.39 variability of the soil type and the strength within the
1.22 1.44 interlayered soil deposits of silty clay and silty sand. To improve
Avg. 1.30 1.42 the accuracy of quality control, a larger container, e.g., 1 l, to be
filled with spoil and a digital balance for weighing spoil has
1052 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053

1 5 6 11 12 17 18 23 24 29 1 2 3 4 12 13 14 21 22 23 1 2 3 4 5 6 7 8 9 10
2 4 7 10 13 16 19 22 25 28 9 10 11 5 19 20 15 29 30 24 61 63 62 65 64 67 66 70 69 68
3 8 9 14 15 20 21 26 27 30 8 7 6 18 17 16 28 27 26 25 20 19 18 17 16 15 14 13 12 11
70 69 63 62 55 54 46 45 38 31 64 63 62 55 56 57 46 40 41 31 58 60 59 55 57 56 52 54 53 51
68 64 61 56 53 47 44 39 37 32 65 70 54 61 58 47 39 45 42 32 21 22 23 24 25 26 27 28 29 30

67 65 60 57 52 48 43 40 36 33 66 69 53 60 59 48 38 44 43 33 41 43 42 45 44 47 46 49 48 50

66 59 58 51 50 49 42 41 35 34 67 68 52 51 50 49 37 36 35 34 40 39 38 37 36 35 34 33 32 31

1 3 4 5 16 17 18 19 20
2 1 2 3 4 5 16 17 18 19 20
12 13 14 15 6 27 28 29 30 21 12 14 13 15 6 30 27 29 28 21
11 10 9 8 7 26 25 24 23 22 11 10 9 8 7 26 25 24 23 22
58 59 60 61 62 35 34 33 32 31 58 59 60 61 62 31 32 33 34 35
57 68 69 70 63 36 47 46 45 44 57 66 69 63 68 44 49 45 47 36
56 67 66 65 64 37 48 49 50 43 56 70 65 67 64 43 50 46 48 37
55 54 53 52 51 38 39 40 41 42 55 54 53 52 51 42 41 40 39 38

Fig. 8. Types of the grout hole layout.

Table 6 overlapping type 6) than perimeter columns (e.g., overlapping

Column overlapping types among various grout hole layouts. type 0). As shown in Fig. 8, five grout hole layouts of 70 jet
grout columns (10 by 7) with a grouting sequence in red lines
Column overlapping type 0 1 2 3 4 5 6 Total Type 4þ5 þ6
are suggested by considering the shortest distance of moving
Layout 1 1 11 21 28 5 4 0 70 9 one grouting machine from hole to hole and to provide the
Layout 2 1 29 11 9 7 7 6 70 20 most infill columns. Among these layouts, grout hole layout 3
Layout 3 4 36 3 0 11 5 11 70 27
gives the largest number of infill columns of overlapping type
Layout 4 1 28 12 9 6 10 4 70 20
Layout 5 1 29 13 8 6 4 9 70 19 4, 5, and 6 in Table 6 and is apparently the smart choice.

proved to be very encouraging in measuring the spoil density
instead of using mud balance (Langhorst et al., 2007).
8. Grout column installation sequence
To minimize the risk of any leakage of water or soil, a jet
grout block must be installed by many overlapping columns
with requisite accuracy. Grouting sequence is usually an
extremely important parameter which has either a direct or
indirect impact on the quality of grouting works (Hurley and
Crockford, 2010). In this study, a total of seven types of
overlapping columns due to grouting sequence are defined, as
shown in Fig. 7. The number of overlapping type indicates that
the next column is adjacent to or is circled by the same number
of columns. The variations of spoil volume and spoil density
for different overlapping types are listed in Table 5. It can be
observed that with higher numbers of overlapped columns,
larger spoil density was noted, although the small sampling
amount for types 5 and 6 may not be sufficiently representa-
tive. On the other hand, the correlation between the volume
ratio of the spoil return to grout and the overlapping type is not
as obvious as to spoil density. Since we already know that the
column density is greater than the spoil density, the grouting
sequence can be rearranged to have more infill columns (e.g.,
9. Concluding remarks
The tunnel boring machine was successfully driven through
the pile foundation below an underground parking building,
with not experience of heave or settlement indicated by the
monitoring results from electronic beam sensors. From the
results and calculations discussed in this paper, the following
conclusions can be drawn:
(1) The control chart of the ratio between spoil return volume
and grout volume can effectively provide an early warning
of sand boiling or groundwater inflow if abnormal frequent
violations of the upper control limit is imminent.
(2) A back calculation from the mean values of the spoil
density and spoil flow rate provides the estimated column
diameter in the field at around 1.56 m, only 2.5% smaller
than the design diameter of 1.6 m.
(3) To improve the measuring accuracy of spoil density, a larger
volume (e.g., 1 l) container and a digital balance are highly
recommended rather than a hydrometer and mud balance.
(4) To optimize the quality of the jet grout column, the
grouting sequence can be arranged in a way that more
infill columns are jet-grouted. The extra time and effort
required to move the grouting machine around also should
be taken into consideration.
 J.C. Ni, W.-C. Cheng / Soils and Foundations 54 (2014) 1039–1053
Acknowledgements
The authors wish to acknowledge the contribution of Mr.
Pu-Yen Chen who has systematically marshaled massive field
records. Thanks are also due to anonymous reviewers for the
valuable comments and corrections to improve this manuscript.
This research would not have been possible without the
financial support from the National Science Council, Taiwan,
under Contract no. NSC96-2622-E-027-019-CC3.
References
Croce, P., Flora, A., 1998. Jet-grouting effects on pyroclastic soils. Riv. Ital.
Geot. 2, 5–14.
Croce, P., Flora, A., 2000. Analysis of single-fluid jet grouting. Geotechnique
50 (6), 739–748.
Croce, P., Modoni, G., 2007. Design of jet-grouting cut-offs. Ground Improv.
11 (1), 11–19.
Flora, A., Lignola, G., Manfredi, G., 2007. A semi-probabilistic approach to
the design of jet grouted umbrellas in tunneling. Ground Improv. 11 (4),
207–217.
Flora, A., Lirer, S.,Monda, M., 2012. Probability design of massive jet grouted
water sealing barriers. In: Fourth International Conference on Grouting and
Deep Mixing, New Orleans, LA.
Hurley, T.M., 2005. Soil Stabilization Through Jet Grouting. North American
No-Dig, Orlando, FL, 2005.
Hurley, T.M., Crockford, R., 2010. Innovative use of jet grouting for earth
retention, underpinning and water control. In: 2010 Earth Retention
Conference, GSP 208, R. Finno, Y.M.A. Hashash,P. Arduino, (Eds.),
American Society of Civil Engineers, Reston, VA., 417-428.
Ho, C.E., 2011. Evaluation of jet grout formation in soft clay for tunnel
excavation. In: Han, J., Alzamora, D.A. (Eds.), Geo-Frontiers 2011:
Advances in Geotechnical Engineering, GSP 211. American Society of
Civil Engineers, Reston, VA (CD-ROM).
JJGA, 2002. Jet Grout Technical Information, 10th ed. Japanese Jet Grout
Association (JJGA).
Juran, J.M., 2010. Juran’s Quality Handbook, sixth ed. McGraw-Hill.
1053
Kauschinger, L.J., Hankour, R., Perry, E.B., 1992. Methods to estimate
composition of jet grout bodies. In: Proceedings of the Conference on
Grouting, Soil Improvement and Geosynthetics, GSP 30, R.H. Borden, R.
O. Holtz, I. Juran, (Eds.), American Society of Civil Engineers, Reston,
VA., 194-205.
Langhorst, O.S., Schat, B.J., de Wit, J.C.W.M., Bogaards, P.J., Essler, R.D.,
Maertens, J., Obladen, B.K.J., Bosma, C.F., Sleuwaegen, J.J., Dekker, H.,
2007. Design and validation of jet grouting for the Amsterdam Central
Station. In: XIV European Conference on Soil Mechanics and Geotechnical
Engineering, Madrid, Spain.
Lignola, G.P., Flora, A., Manfredi, G., 2008. Simple method for the design of
jet grouted umbrellas in tunnelling. J. Geotech. Geoenviron 134 (12),
1778–1790.
Malinin, A., Gladkov, I., Malinin, D., 2010. Experimental research of jet-
grouting parameters in different soil conditions. In: Proceedings of sessions
of GeoShanghai 2010 International Conference, GSP 206, F. Tonon, X.
Liu, W. Wu, (Eds.), American Society of Civil Engineers, Reston, VA, 49-
54.
Miki, G., Nakanishi, W., 1984. Technical progress of the jet grouting method
and its newest type. In: Proceedings of the International Conference on In
Situ Soil and Rock Reinforcement, Paris, pp. 195–200.
Modoni, G., Croce, P., Mongiovì, L., 2006. Theoretical modeling of jet
grouting. Geotechnique 56 (5), 335–347.
Ni, James C., Cheng, W.C., 2012. Characterising the failure pattern of a station
box of Taipei Rapid Transit System (TRTS) and its rehabilitation.
Tunnelling Underground Space Technol. 32, 260–272.
Stark, T.D., Axtell, P.J., Lewis, J.R., Dillon, J.C., Empson, W.B., Topi, J.E.,
Walberg, F.C., 2009. Soil inclusions in jet grout columns. DFI J. 3 (1),
33–44.
Wang, S.Y., Chan, D.H., Lam, K.C., Au, S.K.A., 2013. A new laboratory
apparatus for studying dynamic compaction grouting into granular soils.
Soils Found. 53 (3), 462–468.
Western Electric Company, 1956. Statistical Quality Control handbook, first
ed. Western Electric Co., Indianapolis, IN.
Xu, H.Y., Chen, L.Z., Deng, J.L., 2014. Uplift tests of jet mixing anchor pile.
Soils Found. 54 (2), 168–175.
Yahiro, T., Yoshida, H., Nishi, K., 1982. Soil improvement method utilizing a
high speed and air jet (column jet grout method). In Sixth International
Symposium on Jet Cutting Technology, Surrey, British Columbia.